Machine learning on search sessions to provide intelligent search results

    公开(公告)号:US11886523B2

    公开(公告)日:2024-01-30

    申请号:US17659750

    申请日:2022-04-19

    CPC classification number: G06F16/9535 G06F16/953 G06F16/954 G06N20/00

    Abstract: Techniques for training and/or using a machine learning (ML) algorithm to generate search results are disclosed. An ML algorithm is configured (i) to identify a search starting event and a search terminating event for a search session, where the search session includes multiple navigations across multiple different webpages, and (ii) to derive a dynamic score for the search session, where the dynamic score reflects whether the search session successfully identified an end result that was initially unknown at a time when the search starting event occurred. The trained ML algorithm can then be used during later search sessions to promote better search results and help users identify end targets or results in a faster and more intuitive manner.

    USER INPUT PREDICTIVE ACTION SYSTEM AND METHOD FOR USER INTERFACES

    公开(公告)号:US20230259377A1

    公开(公告)日:2023-08-17

    申请号:US17650743

    申请日:2022-02-11

    CPC classification number: G06F9/45512 G06F3/04842

    Abstract: Systems and methods provide a user input predictive action system and method for an Information Handling System (IHS). In some embodiments, the user input predictive action system includes computer-executable instructions to launch a Graphical User Interface (GUI) and a Command Line Interface (CLI) on a display, store usage information about a plurality of operations performed using at least one of the GUI or the CLI, and identify contextual information about one or more operations performed on the CLI or the GUI using the stored usage information. When the user is using the CLI, the instructions provide the contextual information to the user for aiding the user with entering a command using the CLI or the GUI.

    SMART ONLINE DOCUMENTATION
    4.
    发明公开

    公开(公告)号:US20240020115A1

    公开(公告)日:2024-01-18

    申请号:US17812526

    申请日:2022-07-14

    CPC classification number: G06F8/73 H04L67/306

    Abstract: A system can associate respective tags with respective online documentation content. The system can track user events in interacting with a user interface, wherein the user events are associated with a user account. The system can determine respective scores for respective tags based on the user events and the respective tags. The system can present first online documentation content of the respective online documentation content with the user account based the first online documentation content corresponding to a first tag of the respective tags having a first score of the respective scores that is determined to be above a defined threshold value.

    MACHINE LEARNING ON SEARCH SESSIONS TO PROVIDE INTELLIGENT SEARCH RESULTS

    公开(公告)号:US20230334104A1

    公开(公告)日:2023-10-19

    申请号:US17659750

    申请日:2022-04-19

    CPC classification number: G06F16/953 G06N20/00

    Abstract: Techniques for training and/or using a machine learning (ML) algorithm to generate search results are disclosed. An ML algorithm is configured (i) to identify a search starting event and a search terminating event for a search session, where the search session includes multiple navigations across multiple different webpages, and (ii) to derive a dynamic score for the search session, where the dynamic score reflects whether the search session successfully identified an end result that was initially unknown at a time when the search starting event occurred. The trained ML algorithm can then be used during later search sessions to promote better search results and help users identify end targets or results in a faster and more intuitive manner.

    Promoting APIs Based on Usage
    7.
    发明公开

    公开(公告)号:US20240311881A1

    公开(公告)日:2024-09-19

    申请号:US18183635

    申请日:2023-03-14

    CPC classification number: G06Q30/0283 G06F9/541

    Abstract: A system can assign respective first elements of a first application programming interface (API) to respective first objects in a graph representation, wherein the respective objects correspond to respective syntactic representations of the first API. The system can determine that an action, taken with respect to a user account, has invoked a second API via an API call, wherein respective second elements of the second API are assigned to respective second objects in the graph representation. The system can determine that the first API satisfies a first similarity criterion with respect to the second API. The system can determine that the first API is superior to the second API according to a second criterion. The system can send, for access via the user account, an indication of the second API.

    PEER DISSIMILARITY IDENTIFICATION AND PROFILE GENERATION

    公开(公告)号:US20230368139A1

    公开(公告)日:2023-11-16

    申请号:US17663422

    申请日:2022-05-14

    CPC classification number: G06Q10/101 G06K9/6215 G06Q10/103

    Abstract: Selecting a review panel of dissimilar peers. An asset is reviewed from an ethical perspective by a panel of reviewers. The panel of reviewers includes members that are selected based on their dissimilarity to creators of the asset. Selecting dissimilar members for the panel of reviewers allows bias in the asset to be identified and remedied. A portion of the panel of reviewers may be selected randomly to further improve the effectiveness of the panel of reviewers in reviewing the asset for ethicalness.

    System and method for managing issues based on pain reduction efficiency

    公开(公告)号:US11750479B1

    公开(公告)日:2023-09-05

    申请号:US17863492

    申请日:2022-07-13

    CPC classification number: H04L41/5074 G06F40/30 H04L41/507 H04L41/5067

    Abstract: Methods and systems for managing customer encountered issue resolution are disclosed. To manage the customer encountered issue resolution, tickets for the customer encountered issues may be managed to reduce suffering and reduce the rate of ticket churn. To do so, customer pain levels may be identified and used to prioritize resource deployment for ticket resolution. The tickets may be classified into groups for resolution via automated or human-assisted processes, which may be limited in capacity. To ascertain the pain levels of the customers, communications from the customers may be emotionally analyzed to identify emotions present in the customers. The identified emotions may be used to identify whether performed remediation processes for resolving the customer encountered issues were successful, and to what extent they were successful. Information obtained through previous ticket resolutions may be used to prioritize which approach to use to resolve subsequently encountered tickets for customer encountered issues.

    RECORDING ETHICS DECISIONS
    10.
    发明公开

    公开(公告)号:US20230222513A1

    公开(公告)日:2023-07-13

    申请号:US17647529

    申请日:2022-01-10

    CPC classification number: G06Q30/018

    Abstract: Ethics as a data set interaction variable is disclosed. Ethics are incorporated into the manipulation of data sets. An ethics engine is configured to prompt a user to determine whether manipulations are driven at least in part by ethics. The ethical reasons and ethics labels provided by the user can be recorded in an ethics database. This allows the ethics database to record ethical information and generate recommended manipulations based on ethics.

Patent Agency Ranking